NORTH DAKOTA GAME AND FISH DEPARTMENT Final Report A Two Phase Population Survey of Mussels in North Dakota Rivers Project T-24-R May 1, 2008 – April 30, 2011 Terry Steinwand Director Submitted by Michael G. McKenna Chief, Conservation and Communications Division Bismarck, ND May 2011 A Two Phase Population Survey of Mussels in North Dakota Rivers Final Report Prepared for the North Dakota Game and Fish Department By Dr. Andre DeLorme Department of Biology, Valley City State University, Valley City ND Table of Contents: I. Executive Summary page 1 II. Introduction page 2 III. Methods page 3 IV. Results page 7 V. Discussion page 14 VI. Literature cited page 18 Acknowledgements page 20 Appendix A: List of Sites sampled page 21 with location and date sampled Appendix B: Sheyenne River page 28 Qualitative Data Appendix C: Sheyenne River page 32 Quantitative Data Appendix D: Range Maps of North page 37 Dakota Mussel Species I. EXECUTIVE SUMMARY Freshwater mussel populations are in serious decline in many areas of North America due to pollution, placement of dams, and other factors. The purpose of this project was to document the species found in the state of North Dakota, determine their ranges, and gather information on their population dynamics. In year one of this project (summer of 2008) we conducted a qualitative rapid assessment protocol. Timed searches were performed at 153 sites on 28 rivers; 7,780 mussels were identified and measured with the vast majority being returned to their site of collection. In year two of this project (summer of 2009) we took a subset of 30 sites from the 2008 data and returned for a more rigorous quantitative sampling protocol. Also, qualitative sampling on 7 more sites was done. In year three we performed a combination of qualitative and quantitative sampling to complete our coverage of the state. We sampled an additional 80 qualitative sites and 6 quantitative sites. The majority of these sites, 49, were located in the western part of the state and are part of the Missouri River Drainage. An additional 29 sites were done on rivers in the Red River Basin and 2 sites were done on Long Creek which is a part of the Souris River Drainage. Several points can be made from our three years of data. The western part of the state is relatively devoid of mussels. For example, our qualitative searches of the Little Missouri found no live mussels. The eastern half of the state, on the other hand, exhibited healthy populations of mussels. Specifically the Sheyenne, Maple, and Goose Rivers had good numbers of mussels with diverse populations. We had difficulty trying to sample mussels in the two large rivers, the Red River and the Missouri River. This could be a combination of the inefficiency of our dredge and the difficulty in finding mussels beds in large rivers. We documented two new species records for the state; the deer toe, Truncilla truncata, and the fragile papershell, Leptodea fragilis. This brings the total number of mussel species in the state to 15. 1 II. INTRODUCTION Mussels are one of the most threatened groups of animals in North America. Williams et al. (1993) estimated that 71.1% of the North American mussel fauna was endangered, threatened, or of special concern. A recurring theme of many recent mussel surveys is the decline in both abundance and diversity in the native mussel species. Jones et al. (2001) report that mussels species have been reduced from 27 to 18 species in the upper Clinch River in Tazewell County, Virginia. In a study of the Republican and Smoky Hill river basins in northwestern Kansas, Bergman et al. (2000) found that 11 species have been extirpated when compared to the historical fauna. Poole and Downing (2004) examined mussel demographics in Iowa watersheds comparing data from 1984 and 1985 to a resampling effort done in 1998. They found that maximum species richness for a site was reduced from 22 to 15 species and that all mussel species were extirpated from 47% of the reaches sampled. This particular study is troublesome in that it involves rivers and streams in a primarily agricultural area, and states that a large component of the loss in mussels is due to the agricultural practices in the area. Since North Dakota is largely an agricultural state, it is important to document the current status of our mussel populations. Alan Cvancara, a geology professor at the University of North Dakota, did the majority of work in the state on mussels back in the 1960’s and 1970’s (Cvancara, 1970; Cvancara and Freeman, 1978; Cvancara, 1983). However, very little is known about the current status of North Dakota mussels. The last thorough mussel survey in North Dakota was done over 18 years ago by the ND Game and Fish (Jensen et al., 2001). This study focused exclusively on the Sheyenne and Red Rivers. The Valley City State University (VCSU) Macroinvertebrate lab has done several small scale studies on the Sheyenne and in many places has found healthy mussel populations (Tompkins and DeLorme, 2007; DeLorme, unpublished data). Other rivers in the state have not been surveyed since the 1970’s and most work done was 30 to 40 years ago (Cvancara, 1970; Cvancara, 1983). Taking into consideration the reported changes in mussel populations in other areas, it is important that we document the mussel fauna of North Dakota rivers and streams. Mussels are long-lived organisms that act as good indicators of water quality. It is very important that their distribution and population demographics be documented. 2 III. METHODS III. A. Qualitative sampling Wadeable Rivers Our qualitative sampling consisted of a timed search of designated sites. It was patterned after Villella and Smith (2005). In wadeable areas it consisted of a timed search throughout the site for 2 person hours (p-h). In most cases it was a crew of 4 biologists searching for 30 minutes. The four searchers would spread themselves across the river and slowly walk upstream. Several other crew members assisted with buckets to collect mussels found. There were several variations on this. In several small narrow streams, we would send 2 searchers in one direction and the other two in the other direction. We also had several sites with large numbers of mussels. Because of the time constraints in measuring mussels, we instituted a policy of stopping after 15 minutes (a total of one person hour) if we found over one hundred mussels by that time. The method of searching depended on the river conditions. In shallow, relatively clear water we used visual and tactile observations for finding mussels. Tactile searching can consist of either feeling with the hands or, in deeper waters, detecting mussels with the feet. For many of the sites we did our searches with snorkeling gear. In deeper and more turbid waters foot tactile searching was employed. The majority of sites had these deeper turbid waters and so we used special sampling nets made for us by WildCo. The nets consist of a rectangular net with ¼ mesh with a handle attached. These nets allowed us to dig into the substrate to collect the mussels. During the search all mussels encountered were collected into buckets. At the end of the prescribed time all mussels were identified, measured, and then returned to their habitat. If we encountered specimens that we were unsure of the identity, we saved up to two specimens of that type for later identification in the lab. We also collected empty shells of freshly dead mussels. In this case, a freshly dead mussel means a shell that has a minimum of erosion and still has intact periostracum and nacre. These empty shells serve two purposes; to document species collected at the site and provide specimens for thin sectioning of shells. All shells were placed in bags and labeled. Non Wadeable Rivers In the Missouri River, where depths are often greater than 1.5 meters, we used a dredge employed from a pontoon with a winch and boom. We would do a minimum of three “pulls” per site. This would consist of aligning the pontoon so that it is backing upstream with the dredge being pulled on the front of the pontoon (which is pointing downstream). A pull would be anywhere from 50 – 800 meters. Once the pull was completed, the dredge would be lifted on to the pontoon deck and mussels and shells would be sorted from the debris in the net. After the dredge was checked, it would be 3 lowered back into the water and the pontoon would start the next pull approximately where the previous pull had ended. Anywhere from 3- 5 pulls were done at a sight. III. B Quantitative Survey Our quantitative surveys were based on the work of Strayer and Smith (2003) and Villella and Smith (2005). Data from the rapid assessment described above was used to stratify our sites into either high or low density sites. Villella and Smith (2005) used an arbitrary number of 30 mussels in one p-h from rapid assessment to determine whether a site was considered low or high density. Based on our experience in year one sampling we used the value of 50 mussels in one p-h as a cutoff between low and high density sites. Another feature of this approach is that the two types of sites, low density and high density, are sampled with two different methods, each adapted for the particular type of site (Strayer and Smith, 2003; Villella and Smith, 2005). Low density sites are sampled using an adaptive clustering method and high density sites are sampled using a systematic sampling method as described in Strayer and Smith (2003).